Coder Toolbox MCP Server provides a set of tools to analyze, modify, and manage Java source files within a project, with a focus on integrating code analysis and automated manipulation workflows. The server exposes capabilities to locate Java classes by name and package, create new classes with proper package structure, and modify class bodies by adding, replacing, or deleting content such as fields, methods, and constructors. It also offers a facility to retrieve test execution logs, which can be used to inspect test results and identify failing tests or flaky tests. Use these tools to streamline code maintenance, refactorings, and testing workflows within your Claude-assisted coding sessions.

Available tools include: get_test_execution_logs to fetch logs from your log directory; locate_java_class to search for a class by name with optional source type and package path filtering; create_java_class to scaffold a new Java class within a specified source or test directory and package; class_add_body to insert new content (fields, methods, constructors, etc.) into an existing class; class_replace_body to apply edits to the class body via a list of old/new text replacements (with an optional dryRun); and class_delete_body to remove specific content from a class body (also with an optional dryRun). These tools return structured JSON with results such as file paths and content diffs, enabling easy integration into automated pipelines or Claude-driven interactions.

Prerequisites:

• Node.js (14.x or newer) and npm installed on your system
• Access to the internet to fetch the MCP server package from npm

1. Install Node.js and npm if you haven't already:

   • macOS (with Homebrew): brew install node
   • Windows: download and run the Node.js installer from https://nodejs.org/
   • Linux (Debian/Ubuntu): sudo apt-get update && sudo apt-get install -y nodejs npm

2. Install the MCP server via npx (no global install required) and start using it directly:

   npx -y lamemind-coder-toolbox-mcp-server

   This will initialize the MCP server environment and provide access to the available tools through the MCP interface. If you prefer to pin the version, install via npm and run the server command as documented by the package.

3. Alternatively, install as a local npm package (optional):

   mkdir coder-toolbox && cd coder-toolbox npm init -y npm install lamemind-coder-toolbox-mcp-server

   Start or invoke the server according to the package's CLI guidance

4. Verify installation by listing available commands or running a sample operation as described in the package's usage guide.

Tips and common considerations:

• Ensure you provide correct parameters when locating classes (className is case sensitive, and packagePath is optional but may require sourceType when used).
• When using class_replace_body or class_delete_body, consider performing a dryRun first to preview changes before applying them.
• The tools return JSON results or diffs; integrate these outputs into your IDE or Claude workflows for automated code updates.
• If your project uses a multi-module Maven/Gradle setup, ensure the MCP server has access to the root or appropriate module directories so package paths and sourceType (source vs. test) resolve correctly.
• Environment variables for the tool (if any) can be added via the env block in the mcp_config; consult the package docs for specific VAR_NAME expectations.